5.0 SP 2 changes for some ImplicitBlocks

, ,
1

Imported from previous forum

2

I'm exploring the possibility of generating Google Protocol Buffers code from the Fix repository and have come up with an small issue: I've noticed that in the Fix 4.4 repositoy xml fileset, some components are defined as ImplicitBlock when there should be defined as ImplicitBlockRepeating. In fact, in the Fix 5.0 SP2 xml fileset, many of these have been changed to ImplicitBlockRepeating. Eg. in 4.4 Hop is defined as an ImplicitBlock and in 5.0 SP2 Hop is defined as ImplicitBlockRepeating. (It's also now called HpGrp). So are the 4.4 files in error? If so, should they be updated? Or do you recommend just going to 5.5 SP2? -Greg

3

Greg,

First there is an active group working on Google Protocol Buffers to FIX support - so there will be an official generator made available in the future.

Also, I would recommend should you want to go ahead and build your own that you work from the unified repository format.

With that said - are you using the 2010 edition of the repository?

We did quite a bit of editing of repository data to correct errors in the 2010 edition. If there are still errors you should report them. We track repository errors in Jira.

I would think any worthwhile code generator would need to work for all versions of FIX. Also, as FIX.4.4 is probably the optimal implementation version overall (with exceptions of course and I don’t want to go down this rabbit hole in this thread), the repository needs to be correct. Reporting of issues you identify is greatly appreciated.

4

Hi Jim,

Glad to here that there’s a group working on FIX for PB. I’ll be interested in their output. (No need to reinvent the wheel!)

Yes, I’ve been using the 2010 addition. The issues I had were in the FIX.4.4 files. And the unified format makes better sense. I’ll check it out and be sure to post any errors I come across.

I’ll have to say, it’s pretty cool to generate code from these files. Thanks for all the work that went into it.

-Greg

Greg,

First there is an active group working on Google Protocol Buffers to FIX support - so there will be an official generator made available in the future.

Also, I would recommend should you want to go ahead and build your own that you work from the unified repository format.

With that said - are you using the 2010 edition of the repository?

We did quite a bit of editing of repository data to correct errors in the 2010 edition. If there are still errors you should report them. We track repository errors in Jira.

I would think any worthwhile code generator would need to work for all versions of FIX. Also, as FIX.4.4 is probably the optimal implementation version overall (with exceptions of course and I don’t want to go down this rabbit hole in this thread), the repository needs to be correct. Reporting of issues you identify is greatly appreciated.

5

Greg,

That group the GPB Encoding subgroup of the High Performance Working Group really needs some participation. I think we could also use some help on the generation code.

Do you think you can participate?

Anyone else interested is also encouraged to help out with this work.

Hi Jim,

Glad to here that there’s a group working on FIX for PB. I’ll be interested in their output. (No need to reinvent the wheel!)

Yes, I’ve been using the 2010 addition. The issues I had were in the FIX.4.4 files. And the unified format makes better sense. I’ll check it out and be sure to post any errors I come across.

I’ll have to say, it’s pretty cool to generate code from these files. Thanks for all the work that went into it.

-Greg

Greg,

First there is an active group working on Google Protocol Buffers to FIX support - so there will be an official generator made available in the future.

Also, I would recommend should you want to go ahead and build your own that you work from the unified repository format.

With that said - are you using the 2010 edition of the repository?

We did quite a bit of editing of repository data to correct errors in the 2010 edition. If there are still errors you should report them. We track repository errors in Jira.

I would think any worthwhile code generator would need to work for all versions of FIX. Also, as FIX.4.4 is probably the optimal implementation version overall (with exceptions of course and I don’t want to go down this rabbit hole in this thread), the repository needs to be correct. Reporting of issues you identify is greatly appreciated.

6

I’d be happy to participate. I also have a few other J&J folks who can also join. I’ll give our contact info to Kathleen so she can get us on the proper distribution list.

-Greg

7

Greg,

Also the Repository Working Group just gained new leadership of Mark Sipos of FIX Flyer and Steve Christinson of CameronTEC. We are working on meeting requirements from the Specificiation Working Group and the High Performance Working Group. Perhaps some of you folks (and anyone else on the thread) would like to join the Repository Working Group as well.

http://www.fixprotocol.org/working_groups/repwg

I’d be happy to participate. I also have a few other J&J folks who can also join. I’ll give our contact info to Kathleen so she can get us on the proper distribution list.

-Greg